OCR Further Discrete 2021 November — Question 5 12 marks

Exam BoardOCR
ModuleFurther Discrete (Further Discrete)
Year2021
SessionNovember
Marks12
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicGroups
TypeNon-group structures
DifficultyChallenging +1.8 This is a multi-part game theory question requiring understanding of play-safe strategies, stability conditions, dominance reduction, and linear programming formulation. While the individual concepts are A-level Further Maths standard, the question requires synthesizing multiple game theory techniques across four parts, with part (c) requiring algebraic manipulation as a function of x and part (d) requiring knowledge of LP formulation for mixed strategies—more demanding than typical textbook exercises but still within established FM syllabus content.
Spec7.08a Pay-off matrix: zero-sum games7.08b Dominance: reduce pay-off matrix7.08c Pure strategies: play-safe strategies and stable solutions7.08d Nash equilibrium: identification and interpretation

5 Alex and Beth play a zero-sum game. Alex chooses a strategy P, Q or R and Beth chooses a strategy \(\mathrm { X } , \mathrm { Y }\) or Z . The table shows the number of points won by Alex for each combination of strategies. The entry for cell \(( \mathrm { P } , \mathrm { X } )\) is \(x\), where \(x\) is an integer. \begin{table}[h]
\captionsetup{labelformat=empty} \caption{Beth}
XYZ
\cline { 3 - 5 }P\(x\)32
\cline { 3 - 5 }Q40- 2
\cline { 3 - 5 }R- 3- 1- 3
\cline { 3 - 5 }
\cline { 3 - 5 }
\end{table} Suppose that P is a play-safe strategy.
    1. Determine the values of \(x\) for which the game is stable.
    2. Determine the values of \(x\) for which the game is unstable. The game can be reduced to a \(2 \times 3\) game using dominance.
  1. Write down the pay-off matrix for the reduced game. When the game is unstable, Alex plays strategy P with probability \(p\).
  2. Determine, as a function of \(x\), the value of \(p\) for the optimal mixed strategy for Alex. Suppose, instead, that P is not a play-safe strategy and the value of \(x\) is - 5 .
  3. Show how to set up a linear programming formulation that could be used to find the optimal mixed strategy for Alex.

Question 5:
AnswerMarks Guidance
5(a) (i)
If x  2, row maximin = 2 and col minimax = 2  stable
AnswerMarks
gameM1
A1
AnswerMarks
[2]1.1
1.1Row minima (P = x, 2; Q = –2; R = –3)
Stable when x  2
Allow x > 2
AnswerMarks
(ii)Column maxima are max(x, 4), 3, 2
P is a play-safe strategy  x  -2
AnswerMarks
If -2  x < 2, row maximin = x and col minimax = 2 unstableM1
A1
AnswerMarks
[2]1.1
1.1Column maxima, or equivalent (or implied from part (a)(i))
Unstable when -2  x < 2
Accept -2 < x < 2 but not just x < 2
AnswerMarks Guidance
5(b) X Y Z
P x 3 2
AnswerMarks Guidance
Q 4 0 -2B1
[1]1.1 Row R deleted
5(c) If Beth plays X: Alex expects px + 4(1 – p) or 4 + (x – 4)p
If Beth plays Y: Alex expects 3p
If Beth plays Z: Alex expects 2p – 2(1 – p) or 4p – 2
4p – 2 = px + 4(1 – p)
6
 p =
AnswerMarks
8−𝑥M1
A1
M1
A1
AnswerMarks
[4]3.1a
3.1a
3.2a
AnswerMarks
1.1Calculating expressions for at least two of these
All three correct in any form
Graph is optional and carries no marks
Setting their expression for X = 4p – 2 or = 3p
Correct expression for p
AnswerMarks Guidance
x3 2
5(d) Add 5 (or greater) throughout to make all entries non-negative
0 8 7
9 5 3
2 4 2
Choose row P with probability p, row Q with probability q and row R with
probability r
Maximise M = m – 5
Subject to m  9q + 2r
m  8p + 5q + 4r
m  7p + 3q + 2r
p + q + r  1
AnswerMarks
and m, p, q, r  0B1
B1
B1
AnswerMarks
[3]3.3
3.4
AnswerMarks
3.4Add a constant throughout to make all entries
non-negative
Or using reduced matrix
Need not be stated, may use different notation
e.g. p , p , p
1 2 3
m  ap + bq + cr where a, b, c are the values
from a column of their matrix (or the original
matrix if no evidence of change to matrix)
Or m  ap + bq using reduced matrix
p + q + r  1 (not p + q + r = 1)
Or p + q  1 using reduced matrix
Question 5:
5 | (a) | (i) | Row minima are min(x, 2), -2, -3
If x  2, row maximin = 2 and col minimax = 2  stable
game | M1
A1
[2] | 1.1
1.1 | Row minima (P = x, 2; Q = –2; R = –3)
Stable when x  2
Allow x > 2
(ii) | Column maxima are max(x, 4), 3, 2
P is a play-safe strategy  x  -2
If -2  x < 2, row maximin = x and col minimax = 2 unstable | M1
A1
[2] | 1.1
1.1 | Column maxima, or equivalent (or implied from part (a)(i))
Unstable when -2  x < 2
Accept -2 < x < 2 but not just x < 2
5 | (b) | X Y Z
P x 3 2
Q 4 0 -2 | B1
[1] | 1.1 | Row R deleted
5 | (c) | If Beth plays X: Alex expects px + 4(1 – p) or 4 + (x – 4)p
If Beth plays Y: Alex expects 3p
If Beth plays Z: Alex expects 2p – 2(1 – p) or 4p – 2
4p – 2 = px + 4(1 – p)
6
 p =
8−𝑥 | M1
A1
M1
A1
[4] | 3.1a
3.1a
3.2a
1.1 | Calculating expressions for at least two of these
All three correct in any form
Graph is optional and carries no marks
Setting their expression for X = 4p – 2 or = 3p
Correct expression for p
x | 3 | 2
5 | (d) | Add 5 (or greater) throughout to make all entries non-negative
0 8 7
9 5 3
2 4 2
Choose row P with probability p, row Q with probability q and row R with
probability r
Maximise M = m – 5
Subject to m  9q + 2r
m  8p + 5q + 4r
m  7p + 3q + 2r
p + q + r  1
and m, p, q, r  0 | B1
B1
B1
[3] | 3.3
3.4
3.4 | Add a constant throughout to make all entries
non-negative
Or using reduced matrix
Need not be stated, may use different notation
e.g. p , p , p
1 2 3
m  ap + bq + cr where a, b, c are the values
from a column of their matrix (or the original
matrix if no evidence of change to matrix)
Or m  ap + bq using reduced matrix
p + q + r  1 (not p + q + r = 1)
Or p + q  1 using reduced matrix
5 Alex and Beth play a zero-sum game. Alex chooses a strategy P, Q or R and Beth chooses a strategy $\mathrm { X } , \mathrm { Y }$ or Z . The table shows the number of points won by Alex for each combination of strategies. The entry for cell $( \mathrm { P } , \mathrm { X } )$ is $x$, where $x$ is an integer.

\begin{table}[h]
\begin{center}
\captionsetup{labelformat=empty}
\caption{Beth}
\begin{tabular}{ l c | c | c | c | }
 &  & \multicolumn{1}{c}{X} & \multicolumn{1}{c}{Y} & \multicolumn{1}{c}{Z} \\
\cline { 3 - 5 }
 & P & $x$ & 3 & 2 \\
\cline { 3 - 5 }
 & Q & 4 & 0 & - 2 \\
\cline { 3 - 5 }
 & R & - 3 & - 1 & - 3 \\
\cline { 3 - 5 }
 &  &  &  &  \\
\cline { 3 - 5 }
\end{tabular}
\end{center}
\end{table}

Suppose that P is a play-safe strategy.
\begin{enumerate}[label=(\alph*)]
\item \begin{enumerate}[label=(\roman*)]
\item Determine the values of $x$ for which the game is stable.
\item Determine the values of $x$ for which the game is unstable.

The game can be reduced to a $2 \times 3$ game using dominance.
\end{enumerate}\item Write down the pay-off matrix for the reduced game.

When the game is unstable, Alex plays strategy P with probability $p$.
\item Determine, as a function of $x$, the value of $p$ for the optimal mixed strategy for Alex.

Suppose, instead, that P is not a play-safe strategy and the value of $x$ is - 5 .
\item Show how to set up a linear programming formulation that could be used to find the optimal mixed strategy for Alex.
\end{enumerate}

\hfill \mbox{\textit{OCR Further Discrete 2021 Q5 [12]}}